Semantically generated video method and system

Abstract

A semantically generated video method and system receives a request to generate an imaginative scenario that is embodied by a video and receives a description of the imaginative scenario. The request and description are interpreted by one or more trained computer-implemented neural networks. The one or more trained neural networks are trained using a training set that comprises syntactical elements and images and learn during the training correspondences between each of a plurality of subsets of the syntactical elements and one or more patterns of pixels in the images. Representations of pixels are generated by applying the one or more trained neural networks in accordance with the learned training correspondences and contexts that are associated with the imaginative scenario. A video stream is provided to a user that includes the pixels.

1 . A computer-implemented video generation method, comprising:
 receiving a communication comprising a plurality of syntactical elements, wherein the communication comprises a request that a computer-implemented system generate an imaginative scenario that is embodied by a video, wherein the communication further comprises a description of the imaginative scenario;   interpreting automatically the communication by applying at least onetrained neural networks, wherein the at least one trained neural networks is trained using a training set that comprises training syntactical elements and training images, wherein the at least one trained neural networks learns during training correspondences between each of a plurality of subsets of the training syntactical elements and one or more patterns of pixels in the training images;   generating automatically, by applying the at least one trained neural networks, a representation of a first plurality of pixels that is in accordance with the learned training correspondences and one or more contexts that are associated with the imaginative scenario;   generating automatically, by applying the at least one trained neural networks, a representation of a second plurality of pixels that is in accordance with the learned training correspondences and the representation of the first plurality of pixels; and   providing to a user a video stream that comprises the first plurality of pixels and the second plurality of pixels.   
     
     
         2 . The method of  claim 1 , wherein the description of the imaginative scenario comprises a subset of the plurality of syntactical elements. 
     
     
         3 . The method of  claim 1 , wherein the description of the imaginative scenario comprises one or more images. 
     
     
         4 . The method of  claim 3 , wherein the one or more images are arranged in a sequence within a video. 
     
     
         5 . The method of  claim 1 , wherein generating the first plurality of pixels in accordance with the one or more contexts is performed by applying a plurality of probabilities. 
     
     
         6 . The method of  claim 1 , wherein the first plurality of pixels is included in a first image of the video stream and the second plurality of pixels is included in a second image of the video stream. 
     
     
         7 . The method of  claim 1 , wherein the video stream is further generated in accordance with an inference of a preference of the user that is based on a plurality of usage behaviors that occur prior to receiving the communication. 
     
     
         8 . A computer-implemented system comprising one or more processor-based devices configured to:
 receive a communication comprising a plurality of syntactical elements, wherein the communication comprises a request that a computer-implemented system generate an imaginative scenario that is embodied by a video, wherein the communication further comprises a description of the imaginative scenario;   interpret automatically the communication by applying at least one trained neural network, wherein the at least one trained neural network is trained using a training set that comprises training syntactical elements and training images and the at least one trained neural network learns during training correspondences between each of a plurality of subsets of the training syntactical elements and one or more patterns of pixels in the training images;   generate automatically, by applying the at least one trained neural network, a representation of a first plurality of pixels that is in accordance with the learned training correspondences and one or more contexts that are associated with the imaginative scenario;   generate automatically, by applying the at least one trained neural network, a representation of a second plurality of pixels that is in accordance with the learned training correspondences and the representation of the first plurality of pixels; and   provide to a user a video stream that comprises the first plurality of pixels and the second plurality of pixels.   
     
     
         9 . The computer-implemented system of  claim 8 , wherein the description of the imaginative scenario comprises a subset of the plurality of syntactical elements. 
     
     
         10 . The computer-implemented system of  claim 8 , wherein the description of the imaginative scenario comprises one or more images. 
     
     
         11 . The computer-implemented system of  claim 10 , wherein the one or more images are arranged in a sequence within a video. 
     
     
         12 . The computer-implemented system of  claim 8 , wherein generating the first plurality of pixels in accordance with the one or more contexts is performed by applying a plurality of probabilities. 
     
     
         13 . The computer-implemented system of  claim 8 , wherein the first plurality of pixels is included in a first image of the video stream and the second plurality of pixels is included in a second image of the video stream. 
     
     
         14 . The computer-implemented system of  claim 8 , wherein the video stream is further generated in accordance with an inference of a preference of the user that is based on a plurality of usage behaviors that occur prior to receiving the communication. 
     
     
         15 . A computer-implemented system comprising one or more processor-based devices configured to:
 receive a communication comprising a plurality of syntactical elements, wherein the communication comprises a request that a computer-implemented system generate an imaginative scenario that is embodied by a video, wherein the communication further comprises a description of the imaginative scenario;   interpret automatically the communication by applying at least one neural network, wherein the at least one trained neural network is trained using a training set that comprises training syntactical elements and training images and the at least one trained neural network learns during training correspondences between each of a plurality of subsets of the training syntactical elements and one or more patterns of pixels in the training images;   generate automatically, by applying the at least one trained neural network, a representation of a first plurality of pixels that is in accordance with the learned training correspondences and one or more contexts that are associated with the imaginative scenario;   generate automatically, by applying the at least one trained neural network, a representation of a second plurality of pixels that is in accordance with the learned training correspondences and the representation of the first plurality of pixels; and   provide to a user a video stream that comprises the first plurality of pixels and the second plurality of pixels and that further comprises generated syntactical elements that correspond to the first plurality of pixels and the second plurality of pixels.   
     
     
         16 . The computer-implemented system of  claim 15 , wherein the description of the imaginative scenario comprises a subset of the plurality of syntactical elements. 
     
     
         17 . The computer-implemented system of  claim 15 , wherein the description of the imaginative scenario comprises one or more images. 
     
     
         18 . The computer-implemented system of  claim 15 , wherein generating the first plurality of pixels in accordance with the one or more contexts is performed by applying a plurality of probabilities. 
     
     
         19 . The computer-implemented system of  claim 15 , wherein the first plurality of pixels is included in a first image of the video stream and the second plurality of pixels is included in a second image of the video stream. 
     
     
         20 . The computer-implemented system of  claim 15 , wherein the generated syntactical elements are probabilistically generated by the one or more trained neural networks in accordance with the learned training correspondences.